Granted Patent B2

US 12,536,696 · App. 17/991,610 · Granted Jan 27, 2026

Inventors: Brian Alexander Paden (Scotts Valley, CA); Steven Dean Gottke (Hayward, CA)

Assignee: Summer Robotics, Inc.

G06T7/74G01S17/89G06T2207/30241

Abstract

Claims (112)

1 . A method for sensing objects using one or more processors that are configured to execute instructions, wherein the instructions perform actions, comprising:

providing a sensing system that employs two or more signal beams to scan a plurality of paths across an object, wherein two or more sensors separately detect signals from the two or more signal beams reflected by the object;

determining a plurality of events based on the separately detected signals, wherein the plurality of events include one or more pairs of events that correspond to one or more pairs of sensors;

generating one or more essential matrices based on one or more position characteristics of the two or more sensors, wherein each essential matrix corresponds to each pair of sensors;

comparing the one or more pairs of events associated with the one or more pairs of sensors based on each essential matrix that corresponds to each pair of sensors;

providing one or more error scores for the one or more pairs of events based on the comparison; and

in response to an error score for a pair of events being less than a threshold value, associating each event in the pair of events with a same point location on the scanned object, wherein triangulation of the point location is employed to determine at least a three-dimensional position of the object.

2 . The method of claim 1 , wherein comparing the one or more pairs of events associated with the one or more pairs of sensors, further comprises, substituting a fundamental matrix that corresponds to each pair of sensors for the essential matrix.

3 . The method of claim 1 , wherein determining the plurality of events based on the separately detected signals, further comprises:

providing a time window based on one or more of one or more characteristics of the sensing system or one or more characteristics of the object;

detecting one or more signals that occur within the time window; and

determining the plurality of events based on the one or more detected signals.

4 . The method of claim 1 , further comprising:

generating one or more parametric trajectories based on the plurality of events; and

determining the one or more pairs of events based on the one or more parametric trajectories, wherein each event corresponds to a point on a parametric trajectory.

5 . The method of claim 1 , wherein determining the plurality of events based on the separately detected signals further comprises:

generating a plurality of parametric trajectories based on the detected signals;

associating each parametric trajectory with a separate sensor based on which sensor is associated with the detected signals that correspond to each parametric trajectory;

determining one or more points from each parametric trajectory based on the curve segment that fits the parametric trajectory; and

generating one or more events based on the one or more points.

6 . The method of claim 1 , wherein the sensing system further comprises:

generating the two or more signal beams based on two or more lasers that are included in the sensing system.

7 . The method of claim 1 , wherein the two or more sensors include one or more of an event sensor or an image sensor.

8 . The method of claim 1 , wherein determining the plurality of events based on the separately detected signals, further comprises:

determining a position on a sensor that detects the separately detected signal; and

determining a time that corresponds to the time the sensor detects the separately detected signal.

9 . A system for sensing objects:

a network computer, comprising:

a memory that stores at least instructions; and

one or more processors configured to execute instructions, wherein the instructions perform actions, including:

providing a sensing system that employs two or more signal beams to scan a plurality of paths across an object, wherein two or more sensors separately detect signals from the two or more signal beams reflected by the object;

determining a plurality of events based on the separately detected signals, wherein the plurality of events include one or more pairs of events that correspond to one or more pairs of sensors;

generating one or more essential matrices based on one or more position characteristics of the two or more sensors, wherein each essential matrix corresponds to each pair of sensors;

comparing the one or more pairs of events associated with the one or more pairs of sensors based on each essential matrix that corresponds to each pair of sensors;

providing one or more error scores for the one or more pairs of events based on the comparison; and

in response to an error score for a pair of events being less than a threshold value, associating each event in the pair of events with a same point location on the scanned object, wherein triangulation of the point location is employed to determine at least a three-dimensional position of the object; and

one or more client computers, comprising:

a memory that stores at least instructions; and

one or more processors configured to execute instructions, wherein the instructions perform actions, including, providing one or more portions of the plurality of events.

10 . The system of claim 9 , wherein comparing the one or more pairs of events associated with the one or more pairs of sensors, further comprises, substituting a fundamental matrix that corresponds to each pair of sensors for the essential matrix.

11 . The system of claim 9 , wherein determining the plurality of events based on the separately detected signals, further comprises:

providing a time window based on one or more of one or more characteristics of the sensing system or one or more characteristics of the object;

detecting one or more signals that occur within the time window; and

determining the plurality of events based on the one or more detected signals.

12 . The system of claim 9 , wherein the one or more processors of the network computer are configured to execute instructions, wherein the instructions perform actions further comprising:

generating one or more parametric trajectories based on the plurality of events; and

determining the one or more pairs of events based on the one or more parametric trajectories, wherein each event corresponds to a point on a parametric trajectory.

13 . The system of claim 9 , wherein determining the plurality of events based on the separately detected signals further comprises:

generating a plurality of parametric trajectories based on the detected signals;

associating each parametric trajectory with a separate sensor based on which sensor is associated with the detected signals that correspond to each parametric trajectory;

determining one or more points from each parametric trajectory based on the curve segment that fits the parametric trajectory; and

generating one or more events based on the one or more points.

14 . The system of claim 9 , wherein the sensing system further comprises:

generating the two or more signal beams based on one or more lasers that are included in the sensing system.

15 . The system of claim 9 , wherein the two or more sensors include one or more of an event sensor or an image sensor.

16 . The system of claim 9 , wherein determining the plurality of events based on the separately detected signals, further comprises:

determining a position on a sensor that detects the separately detected signal; and

determining a time that corresponds to the time the sensor detects the separately detected signal.

17 . A network computer for sensing objects, comprising:

a memory that stores at least instructions; and

one or more processors configured to execute instructions, wherein the instructions perform actions, including:

providing a sensing system that employs two or more signal beams to scan a plurality of paths across an object, wherein two or more sensors separately detect signals from the two or more signal beams reflected by the object;

determining a plurality of events based on the separately detected signals, wherein the plurality of events include one or more pairs of events that correspond to one or more pairs of sensors;

generating one or more essential matrices based on one or more position characteristics of the two or more sensors, wherein each essential matrix corresponds to each pair of sensors;

comparing the one or more pairs of events associated with the one or more pairs of sensors based on each essential matrix that corresponds to each pair of sensors;

providing one or more error scores for the one or more pairs of events based on the comparison; and

in response to an error score for a pair of events being less than a threshold value, associating each event in the pair of events with a same point location on the scanned object, wherein triangulation of the point location is employed to determine at least a three-dimensional position of the object.

18 . The network computer of claim 17 , wherein comparing the one or more pairs of events associated with the one or more pairs of sensors, further comprises, substituting a fundamental matrix that corresponds to each pair of sensors for the essential matrix.

19 . The network computer of claim 17 , wherein determining the plurality of events based on the separately detected signals, further comprises:

providing a time window based on one or more of one or more characteristics of the sensing system or one or more characteristics of the object;

detecting one or more signals that occur within the time window; and

determining the plurality of events based on the one or more detected signals.

20 . The network computer of claim 17 , wherein the one or more processors are configured to execute instructions, wherein the instructions perform actions further comprising:

generating one or more parametric trajectories based on the plurality of events; and

determining the one or more pairs of events based on the one or more parametric trajectories, wherein each event corresponds to a point on a parametric trajectory.

21 . The network computer of claim 17 , wherein determining the plurality of events based on the separately detected signals further comprises:

generating a plurality of parametric trajectories based on the detected signals;

associating each parametric trajectory with a separate sensor based on which sensor is associated with the detected signals that correspond to each parametric trajectory;

determining one or more points from each parametric trajectory based on the curve segment that fits the parametric trajectory; and

generating one or more events based on the one or more points.

22 . The network computer of claim 17 , wherein the sensing system further comprises:

generating the two or more signal beams based on one or more lasers that are included in the sensing system.

23 . The network computer of claim 17 , wherein the two or more sensors include one or more of an event sensor or an image sensor.

24 . The network computer of claim 17 , wherein determining the plurality of events based on the separately detected signals, further comprises:

determining a position on a sensor that detects the separately detected signal; and

determining a time that corresponds to the time the sensor detects the separately detected signal.

25 . A processor readable non-transitory storage media that includes instructions for sensing objects, wherein execution of the instructions by one or more processors on one or more network computers performs actions, comprising:

providing a sensing system that employs two or more signal beams to scan a plurality of paths across an object, wherein two or more sensors separately detect signals from the two or more signal beams reflected by the object;

determining a plurality of events based on the separately detected signals, wherein the plurality of events include one or more pairs of events that correspond to one or more pairs of sensors;

generating one or more essential matrices based on one or more position characteristics of the two or more sensors, wherein each essential matrix corresponds to each pair of sensors;

comparing the one or more pairs of events associated with the one or more pairs of sensors based on each essential matrix that corresponds to each pair of sensors;

providing one or more error scores for the one or more pairs of events based on the comparison; and

in response to an error score for a pair of events being less than a threshold value, associating each event in the pair of events with a same point location on the scanned object, wherein triangulation of the point location is employed to determine at least a three-dimensional position of the object.

26 . The media of claim 25 , wherein comparing the one or more pairs of events associated with the one or more pairs of sensors, further comprises, substituting a fundamental matrix that corresponds to each pair of sensors for the essential matrix.

27 . The media of claim 25 , wherein determining the plurality of events based on the separately detected signals, further comprises:

providing a time window based on one or more of one or more characteristics of the sensing system or one or more characteristics of the object;

detecting one or more signals that occur within the time window; and

determining the plurality of events based on the one or more detected signals.

28 . The media of claim 25 , further comprising:

generating one or more parametric trajectories based on the plurality of events, wherein each parametric trajectory is a parametric representation of a one-dimensional curve segment in a three-dimensional space; and

determining the one or more pairs of events based on the one or more parametric trajectories, wherein each event corresponds to a point on a parametric trajectory.

29 . The media of claim 25 , wherein determining the plurality of events based on the separately detected signals further comprises:

generating a plurality of parametric trajectories based on the detected signals, wherein each parametric trajectory is a parametric representation of a one-dimensional curve segment in a three-dimensional space;

associating each parametric trajectory with a separate sensor based on which sensor is associated with the detected signals that correspond to each parametric trajectory;

determining one or more points from each parametric trajectory based on the curve segment that fits the parametric trajectory; and

generating one or more events based on the one or more points.

30 . The media of claim 25 , wherein the sensing system further comprises:

generating the two or more signal beams based on one or more lasers that are included in the sensing system.

31 . The media of claim 25 , wherein the two or more sensors include one or more of an event sensor or an image sensor.

32 . The media of claim 25 , wherein determining the plurality of events based on the separately detected signals, further comprises:

determining a position on a sensor that detects the separately detected signal; and

determining a time that corresponds to the time the sensor detects the separately detected signal.

Assignments (1)

Continuity (3)

References Cited (182)

Cited By (1)

US 12,687,389